Metrics for Matrix-valued Measures via Test Functions

It is perhaps not widely recognized that certain common notions of distance between probability measures have an alternative dual interpretation which compares corresponding functionals against suitable families of test functions. This dual viewpoint extends in a straightforward manner to suggest metrics between matrix-valued measures. Our main interest has been in developing weakly-continuous metrics that are suitable for comparing matrix-valued power spectral density functions. To this end, and following the suggested recipe of utilizing suitable families of test functions, we develop a weakly-continuous metric that is analogous to the Wasserstein metric and applies to matrix-valued densities. We use a numerical example to compare this metric to certain standard alternatives including a different version of a matricial Wasserstein metric developed in [1], [2].

[1]  Michele Pavon,et al.  A Maximum Entropy Enhancement for a Family of High-Resolution Spectral Estimators , 2012, IEEE Transactions on Automatic Control.

[2]  Jim Freeman.  Probability Metrics and the Stability of Stochastic Models , 1991 .

[3]  A. Müller.  Integral Probability Metrics and Their Generating Classes of Functions , 1997, Advances in Applied Probability.

[4]  Zhi-Quan Luo,et al.  Geometric Methods for Spectral Analysis , 2012, IEEE Transactions on Signal Processing.

[5]  Johan Karlsson,et al.  Metrics for Power Spectra: An Axiomatic Approach , 2009, IEEE Transactions on Signal Processing.

[6]  Tryphon T. Georgiou,et al.  On Matrix-Valued Monge–Kantorovich Optimal Mass Transport , 2015, IEEE Transactions on Automatic Control.

[7]  Tryphon T. Georgiou,et al.  Distances and Riemannian Metrics for Multivariate Spectral Densities , 2011, IEEE Transactions on Automatic Control.

[8]  René Vidal,et al.  The alignment distance on Spaces of Linear Dynamical Systems , 2013, 52nd IEEE Conference on Decision and Control.

[9]  Tryphon T. Georgiou,et al.  Matrix-valued Monge-Kantorovich optimal mass transport , 2013, 52nd IEEE Conference on Decision and Control.

[10]  Marc A. Rieffel,et al.  Metrics on State Spaces , 1999 .

[11]  Francesco D'Andrea,et al.  A View on Optimal Transport from Noncommutative Geometry , 2009, 0906.1267.

[12]  Lipeng Ning,et al.  Matrix-valued optimal mass transportation and its applications , 2013 .

[13]  Johan Karlsson,et al.  Uncertainty Bounds for Spectral Estimation , 2012, IEEE Transactions on Automatic Control.

[14]  Michele Pavon,et al.  Hellinger Versus Kullback–Leibler Multivariable Spectrum Approximation , 2007, IEEE Transactions on Automatic Control.

[15]  Tryphon T. Georgiou,et al.  On the Geometry of Covariance Matrices , 2013, IEEE Signal Processing Letters.

[16]  S. Rachev,et al.  The Methods of Distances in the Theory of Probability and Statistics , 2013 .

[17]  C. Villani.  Topics in Optimal Transportation , 2003 .

[18]  Michele Pavon,et al.  Time and Spectral Domain Relative Entropy: A New Approach to Multivariate Spectral Estimation , 2011, IEEE Transactions on Automatic Control.